Biochemical and Biophysical Research Communications 281, 589 –594 (2001)
doi:10.1006/bbrc.2001.4388, available online at http://www.idealibrary.com on
Bracken (Pteridium aquilinum)-Induced DNA Adducts
in Mouse Tissues Are Different from the Adduct
Induced by the Activated Form of the Bracken
Carcinogen Ptaquiloside
R. N. Freitas,* ,† P. J. O’Connor,* A. S. Prakash,‡ M. Shahin,‡ and A. C. Povey* ,§ ,1
*Cancer Research Campaign Carcinogenesis Group, Paterson Institute for Cancer Research, Manchester, M20 9BX United
Kingdom; †School of Nutrition, Federal University of Ouro Preto, Ouro Preto, Minas Gerais, 35400-000, Brazil; ‡National
Research Centre for Environmental Toxicology, 39 Kessels Road, Coopers Plains, Queensland 4108, Australia; and
§School of Epidemiology and Health Sciences, University of Manchester, Manchester M13 9PT, United Kingdom
Received January 22, 2001
Following treatment with bracken fern (Pteridium
aquilinum) extract and bracken spores a number of
DNA adducts were detected by 32P-postlabeling. Three
of these adducts have been described previously
(Povey et al., Br. J. Cancer (1996) 74, 1342–1348) and in
this study, using a slightly different protocol, four new
adducts, with higher chromatographic mobility, were
detected at levels ranging from 50 to 230% of those
previously described. When DNA was treated in vitro
with activated ptaquiloside (APT) and analysed by butanol extraction or nuclease P1 treatment, only one
adduct was detected by 32P-postlabeling. This adduct
was not present in the DNA from mice treated with
bracken fern or spores, suggesting either that bracken
contains genotoxins other than ptaquiloside or that
the metabolism of ptaquiloside produces genotoxins
not reflected by activated ptaquiloside. However, as
the ATP-derived adduct has been detected previously
in ileal DNA of bracken-fed calves, species-specific differences in the metabolism of bracken genotoxins may
exist, thereby leading to differences in their biological
outcomes. © 2001 Academic Press
Key Words: bracken fern; ptaquiloside; DNA adducts;
32
P-postlabelling.
Bracken fern (genus Pteridium) has been described
as one of the most common plants on the planet (Taylor, 1990) and it is the only plant that is known to cause
tumours naturally in animals (Shahin et al., 1999). The
toxicity and carcinogenicity of bracken fern to domestic
and experimental animals has been extensively de1
To whom correspondence should be addressed at School of Epidemiology and Health Sciences, University of Manchester, M13 9PT
UK. Fax: 61 275 5595. E-mail: [email protected].
scribed (Evans and Mason, 1965; Evans, 1984; IARC,
1986; Pamacku and Price, 1969). This plant contains a
number of toxic components, including flavonoid antioxidants such as quercertin and an unstable glycoside,
ptaquiloside (PT), which is thought to be the principal
carcinogenic compound present in bracken (Hirono et
al., 1984a, 1987). Recent work indicates that other
toxic compounds related to ptaquiloside are also
present in bracken fern (Castillo et al., 1998) but in
lower concentrations.
PT is stable at room temperatures but decomposes in
aqueous solutions. Under acidic conditions it undergoes aromatisation by elimination of the glucose to give
pterosin B, whereas under alkaline conditions it is
activated to give a conjugated dienone (activated
ptaquiloside: APT) upon the liberation of the sugar
group (Fig. 1; Ojika et al., 1987)). APT has been shown
to alkylate DNA via a reactive cyclopropyl ring to form
a number of different DNA adducts (Ojika et al., 1987,
1989). The main labile adducts occur at the N-3 of
adenine, and to a minor extent with N-7 of guanine
(Smith et al., 1994b; Kushida et al., 1994). This N3adenine selectivity has also been observed in anticarcinogenic alkylating agents possessing a cyclopropyl ring intermediate (Hurley et al., 1984; Shalder
et al., 1999). Following nuclease P1 enrichment and
32
P-postlabelling, a single adduct, as yet uncharacterised, was found in DNA from the ileum of bracken fed
calves (Prakash et al., 1996) or APT dosed rats (Shahin
et al., 1998), and in DNA treated in vitro with APT
(Smith et al., 1994b). As ptaquiloside-N3 adenine adducts are labile and depurinate within 24 h (Prakash et
al., 1996) they may not be readily detectable in the
postlabeling assay. Previously, it has been demonstrated that administration of either extract or spores
of bracken fern collected in the UK (Pteridium aquili-
589
0006-291X/01 $35.00
Copyright © 2001 by Academic Press
All rights of reproduction in any form reserved.
Vol. 281, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Structures of ptaquiloside, ptaquilosin, activated ptaquiloside (APT), and its interaction with DNA.
num subsp. aquilinum) induced DNA adducts detectable in the upper gastrointestinal tract of BDF1 mice
by using 32P-postlabeling after butanol extraction for
DNA adducts (Povey et al., 1996). However, the adduct
enrichment method used in the former studies (Smith
et al., 1994a; Prakash et al., 1996; Shahin et al., 1998)
differed from that used in the UK study (Povey et al.,
1996) and as these enrichment methods may extract
chemically different adducts (Gupta and Early, 1988) it
was not immediately apparent whether or not APTinduced DNA adducts were present in the DNA isolated from BDF1 mice.
We thus carried out the present work, using both
butanol and nuclease P1 enrichment procedures, to
determine whether the uncharacterised adducts found
in gastrointestinal tissue from BDF1 mice were the
same or different from the one formed by APT in vitro.
MATERIALS AND METHODS
P-␥-ATP (specific activity 7000 Ci/mmol) was purchased from
ICN Pharmaceuticals, Inc. (ICN Biomedicals, Kingston-uponThames, UK). T 4 polynucleotide kinase (T 4-PNK) and calf spleen
phosphodiesterase (CSPE) were both obtained from BoehringerMannheim (Germany). Micrococcal nuclease (MN) and alkaline
phosphatase were obtained from Sigma (Poole, UK). MN and CSPE
were also purchased from Worthington (Lorne Laboratories, Reading, UK). Plastic PEI-cellulose plates were supplied by SchleicherSchuell (Anderman and Co., Kingston-upon-Thames, UK).
32
Preparation of Bracken Samples
An extract of fresh bracken fern (Pteridium aquilinum subsp.
aquilinum) collected in Bellech, Anglesey, UK was prepared (Povey
et al., 1996). Ptaquiloside was isolated and purified as described
elsewhere (Oelrichs et al., 1995) from bracken collected from in
Southeast Queensland.
Treatment of Animals
Treatment of animals with bracken fern extracts and spores from
the UK has been described previously (Povey et al., 1996) and the
DNA obtained in these experiments was used here.
In Vitro Modification of DNA with Activated
Ptaquiloside
Ptaquiloside was activated by incubation with NaOH and used in
vitro to modify calf thymus DNA (Smith et al., 1994a).
Analysis of DNA Adducts
DNA samples from the in vivo experiments were extracted using a
standard phenol-chloroform method after treatment of the tissue
with RNase A and proteinase K (Povey et al., 1996). Both butanol
and nuclease P1 enrichment procedures were used to analyse for
DNA adducts by 32P-postlabelling.
DNA digestion to nucleotides. 10 –25 ␮g of each DNA sample was
digested for 4 h at 37°C using CSPE (0.432 mU/␮g DNA) and MN
(0.495 U/␮g DNA) obtained from Worthington, with 10 mM Tris–
HCl, pH 7.4, 5 mM CaCl 2 and 1 ␮M deoxycorfomycin in a final
volume of 2.5 ␮l/␮g DNA.
Butanol extraction. Aliquots containing 4 ␮g of the DNA digest
were extracted with 1-butanol (Povey et al., 1996) but using 3 rather
than 2 back washes with 1-butanol saturated water.
Nuclease P1 digestion. Aliquots containing 7.89 ␮g of the initial
DNA digest was further incubated with sodium acetate pH 5, zinc
chloride, and nuclease P1 at final concentrations of 40 mM, 0.2 mM,
and 0.31 ␮g/␮l, respectively. The reaction was allowed to proceed for
30 min at 37°C, then stopped by adding 1 ␮l of 0.92 mM Tris and the
samples were dried in vacuo.
590
Vol. 281, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Phosphorimages of butanol-extractable DNA adducts detected in the upper gastrointestinal tract of mice following treatment
with water (A), bracken fern extract (B), and bracken spores (C).
32
P-postlabelling and TLC. The adducted nucleotides were 32Ppostlabelled for 30 min at 37°C in 30 mM Tris-HCl (pH 9.5) containing 10 mM magnesium chloride, 10 mM DTT, 1 mM spermidine, 3.0
␮M ATP (total concentration) using 2.5 units T 4 PNK and 170 ␮Ci
32
P-␥-ATP in a total volume of 10 ␮l. Labelled samples were spotted
on PEI-cellulose TLC plates and chromatographed as previously
described (Povey et al., 1996) but with the plates only run once in D3
and D4. Solvents used for development of the thin layer plates were
as follows: D1, 1.0 M sodium phosphate, pH 6.5; D3, 3.5 M lithium
formate, 8.5 M urea, pH 3.5; D4, 1.2 M lithium chloride, 0.5 M Tris,
8.5 M urea, pH 8; and D5, 1.7 M sodium phosphate, pH 6.
Normal nucleotides and adduct quantitation. The plates were
dried and exposed to a phosphorimager screen for up to 72 h. The
samples were visualised using a Molecular Dynamic Storm 860
image analysis system with ImageQuant software. Adducts were
quantified by comparing the signal with that of a known amount of
the labelling mix spotted onto a TLC plate and exposed together on
the same phosphor screen. To determine the normal levels of nucleotides released, 10 ␮g aliquots of the same DNA digest were treated
with alkaline phosphatase to give nucleosides that were quantified
by HPLC (Cooper et al., 1992).
RESULTS
Upper gastrointestinal DNA samples (oesophagus,
stomach and ileum) from BDF1 mice treated intragastrically with bracken extract or spores contained several adducts (Figs. 2B and 2C) that were not found in
the DNA samples from untreated control animals (Fig.
2A), or in untreated calf thymus DNA used as a negative control (data not shown). The adducts located near
to the origin (Nos. 1, 2, and 3 in Fig. 2C) were described
previously (Povey et al., 1996). In addition to these
adducts, 4 other spots with higher chromatographic
mobility were detected (Nos. 4 –7 in Fig. 2C). Assuming
quantitative recovery, the levels of adducts 4 and 5
were 50 and 90% of adduct 2 respectively whereas
levels of adducts 6 and 7 were, respectively, 160 and
230% higher than the level of adduct 2. Adducts recovered using butanol extraction were similar to those
recovered using nuclease P1 digestion, except for ad-
duct 1, which was poorly recovered after nuclease P1
digestion (Fig. 3).
Only one adduct was detected in calf thymus DNA
reacted in vitro with APT when the sample was treated
with either nuclease P1 APT (Fig. 4) or extracted with
butanol (data not shown). This APT derived adduct
had a chromatographic mobility that was similar to
adducts 6 and 7 but subsequent cochromatography
experiments indicated a chemically different identity
for the APT-derived adduct. Hence, we conclude that
the APT induced DNA adduct was not present in detectable amounts in the DNA obtained from BDF1 mice
treated with bracken fern extract or bracken spores.
DISCUSSION
Previously, the presence of DNA adducts in upper
gastrointestinal tissue of BDF1 mice treated with a
single dose of bracken extracts or spores has been
described (Povey et al., 1996). Using butanol extraction
to enrich the adducted nucleotides in these samples,
three major (Nos. 1, 2, and 3: Fig. 2C) adducts were
found that were not present in untreated animals.
These adducts were not characterised, but their chromatographic mobility was similar to adducts formed in
the same tissue from mice dosed with synthesised compounds containing a cyclopropyl ring (Povey et al.,
1996).
Using a slightly different protocol, not only have the
same spots in these samples been detected, but also at
least 4 other adducts (Nos. 4, 5, 6, and 7) with higher
chromatography mobility (Figs. 2B and 2C) were found
that were not present in the control samples (Fig. 2A).
Procedures that result in the preferential enrichment
of different adduct classes (Reddy and Randerath,
1986; Gupta and Early, 1988) were used to further
characterise these adducts. The DNA samples were
digested to nucleotides and then either extracted with
591
Vol. 281, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Phosphorimages of DNA adducts detected in the upper gastrointestinal tract of mice following treatment with bracken fern
extract (A, C) or bracken spores (B, D). DNA from treated animals was enzymatically digested to nucleotides and then subjected to butanol
extraction (A, B) or further treatment with nuclease P 1 (C, D). The adducts were 32P-postlabelled and chromatographed on PEI-cellulose
plates as described in the text.
1-butanol (Gupta, 1985) or treated with nuclease P1
(Reddy and Randerath, 1986). Most of the adducts
(Nos. 2–7, Fig. 2C) were readily detected by both techniques indicating that these were not sensitive to nuclease P1 digestion unlike certain adducts, notably
those derived from aromatic amines or simple alkylating agents and some cyclic nucleotide adducts (Reddy
et al., 1984; Hemmink et al., 1991; Szyfter et al., 1991;
Vaca et al., 1992).
In order to know if any of the adducts detected in
mouse DNA were identical to that formed by APT in
vitro, a cochromatography experiment was performed
(Fig. 4). This indicated that the principal DNA adduct
formed by APT in vitro was not found in digests of the
upper gastrointestinal tissue DNA of mice treated with
bracken fern extracts or spores. The amount of PT in
the bracken fern extract and bracken spores used in
the mouse experiments was not quantified, but it was
previously calculated that approximately 8 mg of
ptaquiloside was administered to each animal (Povey
et al., 1996) based on a published bracken fern content
(Hirono, 1986). However, as PT levels can depend on a
number of factors including bracken species, freshness,
season and drying conditions (Oelrichs et al., 1995;
Smith et al., 1994b), the amount of PT in these samples
may have been too low to permit the detection of PTderived DNA adducts. In a previous work, we (Shahin
et al., 1998) could detect in ileum tissue a PT derived
DNA adduct from rats dosed intravenously with 3 mg
ptaquiloside weekly for 10 weeks. Although, the absence of PT derived DNA adducts may potentially be
ascribed to insufficient PT in the initial bracken extract, the presence of DNA adducts as detected by
32
P-postlabelling, clearly indicates that the bracken
samples did contain other genotoxins: whether these
are unrelated to PT, or could arise from the metabolism
of PT to forms other than APT remains to be determined. However, as an adduct similar to that arising
592
Vol. 281, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Phosphorimages of high mobility DNA adducts detected in calf thymus DNA modified in vitro with APT (A) and the upper
gastrointestinal tract of mice following treatment with bracken fern extract (B) or bracken spores (C). D and E are respectively 32Ppostlabelled samples of DNA adducts induced by bracken fern extract (as in B) and bracken spores (as in C) cochromatographed with the
adduct from calf thymus DNA modified in vitro with APT (as in A). Circles in D and E locate the APT-derived adduct.
from APT-DNA (Prakash et al., 1996), was found in the
ileum from a calf fed with bracken fern, species specific
differences in PT metabolism may exist. In this regard,
it is of interest to note that in calves, the most important site of tumour occurrence is the ileum, followed by
urinary bladder and mammary gland whereas in mice,
leukaemic, stomach and lung tumours are common but
with ileal tumours being a relatively rare occurrence
(Evans, 1984; IARC, 1985).
In summary, DNA adducts derived from APT were
not detected in tissues from mice treated with
Pteridium aquilinum. The possibility that this is due to
low PT levels in the bracken samples used to treat the
mice cannot be ruled out. However as other DNA ad-
ducts were detected following treatment with
Pteridium aquilinum in mice, it is evident that there
is more than one genotoxic substance is present
in Pteridium aquilinum. Whether these genotoxins
are related to the new PT-like compounds such as
Ptaquiloside Z (Castillo et al., 1998) that are now being
identified remains to be determined.
ACKNOWLEDGMENTS
The authors are grateful to Dr. Barry Smith, AgResearch, NZ, for
his helpful comments; to the CNPq and CAPES of the Brazilian
government and from FAPEMIG from Minas Gerais State government, Brazil, for grants to support R.N.F.; the Cancer Research
593
Vol. 281, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Campaign for long-term funding; and also to the British Council for
facilitating the interchange contacts.
REFERENCES
Castillo, U. F., Ojika, M., Alonso-Amelot, M., and Sakagami, Y.
(1998) Ptaquiloside Z, a new toxic unstable sesquiterpene glucoside from the neotropical bracken fern Pteridium aquilinum
var. caudatum. Bioorg. Med. Chem. 6, 2229 –2233.
Evans, I. A. (1984) Bracken carcinogenicity. In Chemical Carcinogenesis (Searle, C. E., Ed.), pp. 1171–1204. American Chemical
Society Monograph 182, Washington, DC.
Evans, I. A., and Mason, J. (1965) Carcinogenicity activity of
bracken. Nature 208, 913–914.
Gupta, R. C. (1985) Enhanced sensitivity of 32P-postlabeling analysis
of aromatic carcinogen:DNA adducts. Cancer Res. 45, 5656 –
5662.
Gupta, R. C., and Early, K. (1988) 32P-adduct assay: Comparative
recoveries of structurally diverse DNA adducts in the various
enhancement procedures. Carcinogenesis 9, 1687–1693.
Hemminki, K., Szyfter, K., and Kadublar, F. F. (1991) Quantitation
of the 32P-postlabeling reaction using cyclic N1,N 2 and C8 modified deoxyguanosine-3-monophosphates as substrates. ChemBiol. Interact. 77, 51– 61.
Hirono, I. (1986) Carcinogenicity of plant constituents: Pyrrolizidine
alkaloids, bracken fern. In Genetic Toxicology of the Diet
(Knudsen, I., Ed.), pp. 45–53. Alan Liss, New York.
Hirono, I., Yamada, K., Niwa, N., Shizuri, M., Ojika, M., Hasaka, S.,
Yamaji, T., Wakamatsu, K., Kigashi, H., Nuyama, K., and
Uosakki, Y. (1984a) Separation of carcinogenic fraction of
bracken fern. Cancer Lett. 21, 239 –246.
Hirono, I., Aiso, S., Yamaji, T., Mori, H., Yamada, K., Niwa, H.,
Ojika, M., Wakamatsu, K., Kigoshi, H., Niiyama, K., and
Uosaki, Y. (1984b) Carcinogenicity in rats of ptaquiloside isolated from bracken. Gann 75, 833– 836.
Hirono, I., Ojino, H., Fujimoto, M., Yamada, K., Yoshida, Y., Igawa,
M., and Okumura, M. (1987) Induction of tumours in ACI rats
given a diet containing ptaquiloside, a bracken carcinogen.
J. Natl. Cancer Inst. 79, 1143–1149.
Hurley, L. H., Reynolds, V. L., Swenson, D. H., Petzold, G. L., and
Scahill, T. A. (1984) Reaction of the antitumor antibiotic CC1065 with DNA: Structure of a DNA adduct with DNA sequence
specificity. Science 226, 843– 844.
IARC (1986) Bracken fern (Pteridium aquilinum) and some of its
constituents. In Monographs on the Evaluation of Carcinogenic
Risk to Humans. No. 40. pp. 47– 65. WHO, IARC, Lyon.
Kushida, T., Uesugi, M., Sugiura, Y., Kigoshi, H., Tanaka, H., Hirokawa, J., Ojika, M., and Yamada, K. (1994) DNA damage by
ptaquiloside, a potent bracken carcinogen: Detection of selective
strand breaks and identification of DNA cleavage products.
J. Am. Chem. Soc. 116, 479 – 486.
Oelrichs, P. B., Ng, J., and Bartley, J. (1995) Purification of
ptaquiloside, a carcinogen from Pteridium aquilinum. Phytochemistry 40(1), 53–56.
Ojika, M., Wakamatsu, K., Niwa, H., and Yamada, K. (1987)
Ptaquiloside, a potent carcinogen isolated from bracken fern
Pteridium aquilinum var. laticulum. Structure elucidation
based on chemical and spectral evidence and reactions with
amino acids, nucleosides and nucleotides. Tetrahedron 43,
5261–5274.
Ojika, M., Sugimoto, K., Okazaki, T., and Yamada, K. (1989) Modification and cleavage of DNA by ptaquiloside. A new potent
carcinogen isolated from bracken fern. Chem. Soc. Chem. Commun., 1775–1777.
Pamacku, A. M., and Price, J. M. (1969) Induction of urinary bladder
cancer in rats by feeding bracken fern (Pteris aquilina). J. Natl.
Cancer Inst. 43, 275–281.
Prakash, A. S., Pereira, T. N., Smith, B. L., Shaw, G., and Seawright,
A. A. (1996) Mechanism of bracken fern carcinogenesis: Evidence for H-ras activation via initial adenine alkylation by
ptaquiloside. Nat. Toxins 4, 221–227.
Povey, A. C., Potter, D., and O’Connor, P. J. (1996) 32P-postlabeling
analysis of DNA adducts formed in the upper gastrointestinal
tissue of mice fed bracken extract or bracken spores. Br. J.
Cancer 74, 1342–1348.
Reddy, M. V., and Randerath, K. (1986) Nuclease P1-mediated enhancement of sensitivity of 32P-postlabeling test for structurally
diverse DNA adducts. Carcinogenesis 7, 1543–1551.
Reddy, M. V., Gupta, R. C., Randerath, E., and Randerath, K. (1984)
32
P-postlabeling test for covalent DNA binding of chemicals in
vivo: Application to a variety of aromatic carcinogens and methylating agents. Carcinogenesis 5, 231–243.
Shahin, M., Smith, B. L., Worral, S., Moore, M. R., Seawright, A. A.,
and Prakash, A. S. (1998) Bracken fern carcinogenesis: Multiple
intravenous doses of activated ptaquiloside induce DNA adducts, monocytosis, increased TNF␣ levels and mammary gland
carcinoma in rats. Biochem. Biophys. Res. Commun. 244, 192–
197.
Shahin, M., Snith, B. L., and Prakash, A. S. (1999) Bracken carcinogens in the human diet. Mutat. Res. 443, 69 –79.
Shalders, R. L., Blanch, G., Brown, C. L., Young, D. J., and Prakash,
A. S. (1999) Stability and DNA alkylation rates of simplest
functional analogs of CC-1065, p-hydroxy and p-amino phenethyl bromides. Chem.-Biol. Interact. 117, 83–94.
Smith, B. L., Shaw, G., Prakash, A. S., and Seawright, A. A. (1994a)
Studies on DNA adduct formation by ptaquiloside, the carcinogen of bracken ferns (Pteridium spp.). In Plant Associated Toxins (Colgate, S. M., and Dorling, P. R., Eds.), Chap. 31, pp.
167–172. CAB International, Wallingford, UK.
Smith, B. L., Seawright, A. A., Ng, J. C., Hertle, A. T., Thomson,
J. A., and Bostock, P. D. (1994b) Concentration of ptaquiloside,
a major carcinogen in bracken fern (Pteridium spp.), from eastern Australia and from a cultivated worldwide collection held in
Sydney, Australia. Nat. Toxins 2, 347–353.
Szyfter, K., Hemminki, K., Crane, A. E., and Watson, W. P. (1991)
Quantitative and kinetic examination of 32P-postlabeling of
etheno-substituted nucleotides. Chem.-Biol. Interact. 80, 99 –
107.
Taylor, J. A. (1990) The bracken problem: A global perspective. AIAS
Occasional Publ. 40, 3–19.
Vaca, C. E., Vodicka, P., and Hemminki, K. (1992) Determination of
malonaldehyde-modified 2-deoxyguanosine-3⬘-monophosphate
and DNA by 32P-postlabeling. Carcinogenesis 13, 593–599.
594